On Mon, May 14, 2018 at 7:43 AM, Dongsheng Yang <dongsheng.yang@xxxxxxxxxxxx> wrote: > > > On 05/11/2018 09:09 PM, Ilya Dryomov wrote: >> >> On Fri, May 11, 2018 at 9:16 AM, Dongsheng Yang >> <dongsheng.yang@xxxxxxxxxxxx> wrote: >>> >>> >>> On 05/10/2018 10:56 PM, Ilya Dryomov wrote: >>> >>> On Tue, May 8, 2018 at 8:13 AM, Dongsheng Yang >>> <dongsheng.yang@xxxxxxxxxxxx> wrote: > > ... >>> >>> Customers complained times >>> >>> "Just tell me what happened, rather than suspend everything and forbid to >>> cancel." >>> >>> So we want to just return error to user and allow them to do a clean work >>> and recovery >>> their containers. >> >> Right, so you are using pool quotas. Support for pool quotas is >> a relatively recent addition (since 4.7), so you are hitting all the >> sharp edges. This is exactly how it works in userspace -- no changes >> were made. >> >> Hanging processes in D state is definitely undesirable, but I'm curious >> how would the recovery look like? There is nothing that can done on the >> customer side once the pool is marked full. It will remain full until >> an administrative action is taken. > > > We can remove some unused rbd images in this case, > > https://github.com/ceph/ceph/pull/12627 > > In our using, there is a reset button at web for each container. When the > user > found no space left, they can remove some unused volumes. Then > we can click the reset button of each container which will reset the > container status, actually, it just stop and then start the container again. > And then this container would be running again. I see, removing an image is one example. I'd be fine with adding an option to return -ENOSPC, off by default. Josh, any objections? Your patch that made blocking optional was dropped, and I'm trying to remember why. Was it just because the OSDs had to be changed to drop such writes on the floor, the problem being that someone with an old osdmap wouldn't know whether to resend? Thanks, Ilya -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
